The sample size to determine the mean of a continuously distributed variable in a population of infinite size may be calculated using the following formula:
Where:
| z | The critical value obtained from the standard normal distribution. For each level of confidence there is a corresponding value of z. The levels of confidence frequently used in biological studies are 90%, 95%, and 99%. The corresponding z values are 1.64, 1.96, and 2.58 respectively. |
| de | The standard deviation of the variable of interest. |
| d | The maximum absolute difference between the sample estimate and the unknown population value. |
Sample size may be adjusted (na) according to the size of the study population (N) as follows:
Where:
| n | The sample size calculated for an infinite population. |
| N | The size of the population under study. |
| Desired precision | For example: to estimate bodyweight in a mob of calves to within 20 kg of the actual value, the absolute precision is ± 20. Acceptable values: any positive number. |
| Level of confidence | The confidence that the user wants to have in the results. Acceptable values: 90%, 95% or 99%. |
| Standard deviation | If the standard deviation is unknown an approximate value can be estimated from the likely range of possible values. For example: if the range of bodyweights in a mob of bulls is estimated to be between 300 kg and 420 kg the range of weights is (420 - 300) = 120 kg. The standard deviation equals one-sixth of this value, 20 kg. Acceptable values: any positive number. |
| Population size | The number of individuals that comprise the population under investigation. Acceptable values: any positive integer. |
|
|
Instituto Nacional de Tecnología Agropecuaria |
EpiCentre, IVABS, Massey University |
